基于决策变量分组的粒子群算法求解大规模优化问题

Particle swarm optimization algorithm based on grouping of decision variables to solve large-scale optimization problems

针对社会学习粒子群算法在求解大规模优化问题时存在的收敛速度慢以及种群多样性缺失等问题,提出一种基于决策变量分组的粒子群算法.根据决策变量间的相关性对决策变量分组,提高算法的收敛速度.采用反向学习策略,通过生成反向解,提高算法的全局寻优能力.采用CEC2010测试函数集对本文算法进行测试,仿真结果与已有典型算法进行对比,验证了本文算法的有效性.

Aiming at the problems of slow convergence speed and lack of population diversity of social learning particle swarm optimization algorithm when solving large -scale optimization problems,a particle swarm algorithm based on grouping decision variables is proposed.The decision variables are grouped according to the correlation between the decision variables,and the convergence speed of the algorithm is improved.The opposition learning strategy is used to improve the global optimization ability of the algorithm by generating opposition solutions.The CEC2010 test function set is used to test the proposed algorithm,and the simulation results are compared with existing typical algorithms to verify the effectiveness of the proposed algorithm.